The email sat in Amelia’s inbox, unread, for an hour. She kept clicking over to it, then retreating back to her blank document. Her breath hitched slightly each time the subject line, “Advisor Departure,” caught her eye. She drummed her fingers against the desk, the rhythm erratic and fast. Finally, with a sigh that felt heavy in her chest, she clicked it open.

The sentences blurred before her eyes. Dr. Chen was leaving. Leaving! To take a position at a university across the country. Amelia felt a coldness creep into her extremities. The carefully constructed research plan, the projected graduation date – all of it seemed to waver, as if seen through heat.

She reread the email, her gaze darting from word to word, searching for some loophole, some error. There wasn’t one. Her stomach churned. She pushed her chair back, the squeak of the wheels echoing in the sudden quiet of her office. She needed air.

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

This story is one of 1,000 stories generated for the emotion worried. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 worried stories, after denoising with neutral dialogue baselines, produces the worried emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the worried vector is projected through the unembedding matrix.

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
🎉-0.233
happy-0.231
"-0.228